Table 4 Results for the performance of the CNN on the five classes patch-level classification task. The performance is evaluated with Cohen’s κ-score, reporting the average and the standard deviation of the models involved in the k-fold cross-validation. The performance is reported for the CNNs trained using the automatically generated weak labels (SKET labels) and the manually created weak labels (GT labels).

From: Unleashing the potential of digital pathology data by training computer-aided diagnosis models without human annotations